MasterPrompt VaultBY MASTERESTAURANT

AI assistant · N5 Complete system

◆ Included in the annual plan💬 AI Community Managerv1.1.0 · 2026-08-27
📷 Works best with an AI that sees photos (invoices, dishes, screens) · optional

Automated Reply System Builder for Restaurant Social Media

Choose at the top where the message lands: a direct message or a public comment. In this one —the direct message— you get the system deciding what a template answers and what goes to a person: classified intents, a response window per time band, a signature and the handover line.

N5
Depth
14/23
Your data
3
Compatible AIs
6
Steps to use it
Depth of the resulttap a level to see what you get
◀ Put out today's fireRun the year ▶
🏛️

N5Complete system

Several chained pieces you install and operate over time, with their review cadence. It changes how the business runs, not one decision.

◆ This assistant 110 of the 164 published assistants are at this level

Is this for you?

What it does and what you get

  • Automating replies is cheap until the day a template answers someone who is hurt: that is when a complaint becomes a screenshot doing the rounds.
  • This assistant doesn't write a channel's replies —its three sister pieces cover public comments, direct messages and frequently asked questions—: it builds the law all three obey.
  • It labels incoming messages by intent, sets how long each may take by daypart, writes the signature and the tone, and above all draws the border: what answers itself and what switches off so a person replies, with an owner and a clock.
  • What comes out can be run by someone three days into the job.

What you end up with

The intent matrix with one row per incoming type —fact question, buying intent, praise, complaint and spam—, its telltale signal, its maximum time and who signs; the one-page common law with signature, openings, closings, window per daypart and the list of what never ships automatically; and the escalation board with its triggers, its recipient, its clock and what gets logged.

EXAMPLE This is what you getAn example built with Sazón de Origen, the house's fictional restaurant.Not your data: your result is built with YOUR restaurant's.See the full example →

Sample result

Sazón de OrigenFictional contemporary Colombian restaurant in Medellín, 2 locations, 6 years in business. Every sample result in the library comes from this same case. Fictional restaurant

The owner asked this: «We're Sazon de Origen, contemporary Colombian cooking in Medellín: two locations, 24 people, a 48.000 COP average check. Between Instagram and WhatsApp about 60 conversations land daily and three of us answer them, each in our own way. I turned on automated replies a month ago and last week a lady complained about a delivery that arrived cold: she got the welcome greeting. The screenshot is going around. I sell in the dining room, through my own delivery and one delivery platform, and I'm after taking operating margin from 8% to 14% without raising menu prices.»

IntentTelltale signalTemplate or personMaximum timeWho signs
Fact question«what time do you open?», «is there parking?», «do you deliver to Envigado?»Template alone15 minutes during service · 12 hours once closedLocation manager
Buying intentasks about a table, party size, or the price of a dishTemplate with a data point: seats, hour and menu price10 minutes between 11:00 and 14:30Location manager
Praisethanks you, tags the photo or names a dish: «the house Bandeja tasted like my mother's»Template plus one handwritten line about THAT dish24 hoursWhoever is on shift
Complaintmentions money, a cold delivery, health, or a named team memberA person, always; the template switches off30 minutes in public · settled privately the same dayDelivery lead or the owner
Spam or pitchagency link, follower promises, the same message repeatedNot answered: archived—Nobody

The row that moves the business is the second one: buying intent during Tuesday-to-Friday lunch carries the shortest clock in the house, because an unanswered «any tables?» is two covers walking somewhere else. Figures are illustrative: the times come from one real week of your own volume, not from an internet average.

This is 2 of 3 parts. Behind the wall sits **the escalation board**: the five triggers —money, health, a named employee, the discount and the thread that already grew— with their recipient, their maximum time and what gets logged for each; plus the COP arithmetic explaining why buying intent runs against 10 minutes while praise can wait 24 hours.

See the full example — free account

Works with these AIs

Claude — the best for this assistant It holds the border without softening it: told that complaints never enter the template lane, it doesn't slip them back in one row down. Paste the whole assistant and have one real week of messages at hand.

How to use it

Choose at the top where it lands. A comment is answered for the people reading and a direct message for whoever wrote: not the same text, and the chip takes you to the right one.
Fill in My Restaurant with your active networks, sales channels, strong dayparts and tone: the system is calibrated on that data, not on some generic restaurant.
Pick what you need on the card —where the message lands (a public comment, a direct message, a frequently asked question, a chained flow…)— and the assistant reframes itself: you don't need another prompt.
Copy the whole assistant and paste it into the AI you already use.
Answer its questions one item at a time: who replies today, from which account, during which hours.
Get the intent matrix, the common law and the escalation board; argue row by row which ones really get automated.

Which of your data it uses

14
of 23

This assistant works with 14 of the 23 fields in «My Restaurant», among them Restaurant name, Business type and City and country. It does not ask out of curiosity: these are what make the answer speak about YOUR scale instead of an industry average — the same calculation on a business of another size returns a number you cannot decide with. You fill them once and they apply across the library, so the second assistant you open already starts with them in place. Whatever you leave blank, the assistant asks for one thing at a time instead of inventing it, and says so before answering, so you know what it is working from.

Restaurant nameBusiness typeCity and countryYour main guestActive social channels (with the @ or the link)Active sales channelsStrong and slow daypartsTeam sizeCommunication toneWhat makes you differentBest-selling dishesAverage check (with currency)Menu price rangeMain goal this year

Always up to date

Every assistant carries a visible version and date. When an AI changes how it works, the assistant's version goes up and you see what changed.

Same method, another variant

Inside the app this is a single control: you press it and the assistant reframes itself, including the variants that don't appear here because they don't have their own page yet.

Where it lands

More assistants in this category

Included in the always-growing library

Access to every published assistant, adapted to your AI and personalised with your restaurant's data, updates included.

Automated Reply System Builder for Restaurant Social MediaN5 · Complete system · Included in the annual plan See the plan

Exclusively for restaurant leaders

AI Executive · Artificial Intelligence for Restaurants

Make 1 person produce like 5. In 8 weeks you turn Artificial Intelligence into a new productivity department inside your restaurant: content, sales, finance, manuals, automations and specialized assistants.

I want to implement AI in my restaurant →
NEW AI Executive — executive Artificial Intelligence program for restaurant leaders